RNA interference is mediated by small interfering RNAs produced by members of the ribonuclease III (RNase III) family, including Dicer. For mechanistic studies, bacterial RNase III has been a valuable model system for the entire family. Previously, we have shown how the dimerization of the endonuclease domain of the enzyme creates a catalytic valley where two catalytic sites are located, how the catalytic valley accommodates a dsRNA in a manner such that each of the two RNA strands is aligned with one of the two catalytic sites, how the hydrolysis of each strand involves both subunits (residues from one subunit are involved in the selection of the scissile bond, while those from the partner subunit are involved in the cleavage chemistry), and how RNase III uses the two catalytic sites to create the 2-nucleotide 3' overhangs in its products. Recently, we have demonstrated how Mg2+ is essential for the formation of a catalytically competent protein-RNA complex, how the use of two Mg2+ ions can drive the hydrolysis of each phosphodiester bond, and how conformational changes in both the substrate and the protein are critical elements for assembling the catalytic complex. Moreover, we have modeled a protein-substrate complex and a protein-reaction intermediate (transition state) complex in a meaningful way. Together, the models and crystal structures suggest a stepwise mechanism for the enzyme to execute the phosphoryl transfer reaction. The structural information of protein-dsRNA interactions and the mechanism of dsRNA processing by bacterial RNase III can be extrapolated to other family members, including eukaryotic Rnt1p, Drosha and Dicer. The folate and shikimate pathways are essential for microorganisms and some of the enzymes in the two pathways are absent from mammals, offering ideal targets for the development of novel antimicrobial agents. For example, the molecular targets for both sulfonamides and trimethoprim are folate pathway enzymes. We have obtained a sufficient amount of structural information for 6-hydroxymethyl-7,8-dihydropterin pyrophosphokinase (HPPK) and dihydroneopterin aldolase (DHNA) in the folate pathway and of shikimate kinase and shikimate dehydrogenase in the shikimate pathway, which allowed us to derive the catalytic mechanism for these enzymes. These enzymes are not targets for any existing drugs and therefore are ideal targets for structure-based design of novel antibiotics. Glutathione S-transferase (GST) catalyzes glutathione conjugation with electrophilic compounds. In preneoplastic and neoplastic cells, specific forms of GST are expressed at high levels and to participate in the cells' resistance to anticancer drugs. Class pi GST (GSTP) is of particular importance in biological resistance to alkylating agents. A new family of GST-activated prodrugs has shown great potential, which function by releasing nitric oxide inside cancer cells. We have achieved GSTP specificity of a lead compound with two structural modifications. In addition, we have determined several GSTP structures containing inactivated glutathione molecules for structural characterization of GSTP in complex with prodrug molecules.